soil erosion and risk assessment for on- and off-farm ... · the area are discussed by boardman...
Post on 25-Aug-2020
3 Views
Preview:
TRANSCRIPT
1
Soil erosion and risk assessment for on- and off-farm impacts: a test case using the Midhurst area, West Sussex, UK
John Boardmanab*, Mark L. Shephearda, Edward Walkera and Ian D.L. Fostercd
a Environmental Change Institute, Oxford University Centre for the Environment, Dyson Perrins Building, South Parks Road, Oxford OX1 3QY, UK
bDepartment of Environmental and Geographical Science, University of Cape Town, Rondebosch 7701, South Africa
cDepartment of Molecular and Applied Biosciences, School of Biosciences, University of Westminster, Cavendish Campus, 115 New Cavendish Street, London W1W 6UW, UK
d Department of Geography, Rhodes University, PO Box 94, Grahamstown 6140, Eastern Cape, South Africa,
Abstract Soil erosion on agricultural land is a growing problem in western Europe and constitutes a threat to soil quality and to the ability of soils to provide environmental services. The off-site impacts of runoff and eroded soil, principally eutrophication of water bodies, sedimentation of gravel-bedded rivers, loss of reservoir capacity, muddy flooding of roads and communities, are increasingly recognised and costed. The shift of funding in the European Union (EU) from production-related to avoidance of pollution and landscape protection, raises issues of cross-compliance: public support for agriculture has to be seen to give value-for-money. In this context risk-assessment procedures have been introduced to help farmers recognise sites where either certain crops should not be grown or anti-erosion measures are required. In England, Defra (2005a) sets out a system of risk-assessment, including ranking of crops susceptible to erosion and anti-erosion measures, that may be selected. We assess this system using field data for an area of erodible soils in the Rother valley, Sussex. The Defra approach correctly identifies most at-risk fields and, taken together with land-use maps, allows non-compliance with advice to be highlighted. We suggest a simple extension to the system which would further identify at-risk fields in terms of possible damage to roads and rivers from muddy runoff. The increased risk of erosion in the study area is associated with certain crops: potatoes, winter cereals, maize and grazed turnips and seems unlikely to be the result of changes in rainfall which over the last 130 years are minimal. We have not evaluated proposed anti-erosion measures in the area because few have been put into practice. The European Water Framework Directive will increasingly focus attention on agricultural fields as a source of river pollution. Assessing the risk of erosion and the need for field testing of suggested approaches, are not simply issues for the EU, but for the management of global agricultural systems. Keywords: Erosion; Risk-assessment; Off-site impacts; Climate change; Agricultural change; Land-use change; Muddy floods -------------------
*Corresponding author. Tel.: +44 1865285183; fax: +44 1865275850. E-mail address: john.boardman@eci.ox.ac.uk (John Boardman)
2
1. Introduction
All agricultural systems are threatened to some degree by erosion. The threat may be over short
time periods or longer; the latter are particularly difficult for farmers to take account of since other
threats to their livelihoods, especially farming economics, are often more immediate. In the
Developed World, although starvation or eviction may not be serious threats, the issue of economic
insolvency is real. Prices change by the year and failure to respond may lead to financial disaster.
For this reason the threat of erosion may be considered to be of minor account especially as the
costs of off-site damage tend to be borne by others or by society at large. Over-production in the EU
has led to attempts to sustain rural livelihoods and shift production-oriented subsidies to support for
environmentally-friendly farming or landscape management. In this context, many EU countries
introduced support for schemes that would reduce phosphorous and nitrate pollution and control
muddy flooding. In England and Wales, one aspect of this policy is the Defra (2005a) advice for
controlling erosion. Farmers are now required to keep land in ‘Good Agricultural and
Environmental Condition’ in order to receive their area-based Single Farm Payment – part of the
so-called ‘Cross-Compliance’ regime (Defra, 2006a). One aspect of the regime requires all farmers
to have carried out a Soil Protection Review, which assesses the risk of erosion on their fields and
requires them to reduce that risk to acceptable levels if they are to continue to receive the Single
Farm Payment. The risk-assessment approach discussed in this paper is thus linked to
environmental payments available to all British farmers (Defra, 2005b). It is therefore a major
financial tool for the achievement of environmental targets.
The Defra assessment of risk of erosion procedures have to be easily understood and implementable
by all farmers and useable on a field-by-field basis. Traditional approaches have involved forms of
modelling which, in some circumstances, are too complex or too data-hungry to be suitable. The
need to return to rather simpler approaches is argued in some quarters (Favis-Mortlock et al., 2001).
3
The question remains as to whether such an approach can give adequate results in terms of
predicting the fields on which, in actuality, erosion occurs.
The aims of this paper are to describe and explain erosion in an area of the Rother valley, West
Sussex, and to use this field-based data to evaluate the Defra scheme for erosion-risk assessment.
Modifications to the scheme will also be considered.
2. The study area
2.1. Lower Greensand soils and erosion risk
This study focuses on an area of soils developed in Lower Greensand beds of Cretaceous age
forming a small but important component of the agricultural soils of southern England (Figure 1).
Their easily worked, sandy and loamy character make them attractive for a wide variety of crops
and this is reflected in a long history of occupation and agriculture (e.g. Scaife and Macphail, 1983).
Figure 1 hereabouts
Several studies on these soils point to the risk of erosion when used for arable farming. In the Silsoe
area of Bedfordshire, erosion was recorded on experimental plots (Morgan, 1977; Morgan et al.,
1987) and the same area was used for the development of the EUROSEM model (Quinton and Catt,
2004; Morgan et al., 1998). Photographic surveys of erosion were conducted by Jackson (1986) and
the area was included in a MAFF erosion monitoring project 1989-94 (Chambers et al., 1992;
Chambers and Garwood, 2000). In the Isle of Wight, Lower Greensand soils of the Arreton valley,
have been the subject of two studies (Turnbull, 1985; Evans, 1993) and off-site impacts on roads in
the area are discussed by Boardman (1994). At Hascombe, Surrey, and Albourne and Washington
in West Sussex, serious erosion was recorded on fields of winter cereals, strawberries and post-
harvest maize, respectively (Boardman, 1983a and b; Boardman, unpublished).
4
2.2. The Rother Valley
The area of the Lower Greensand outcrop between Petersfield and Petworth is drained by the east
flowing River Rother (Figure 2). Land adjacent to the river is mostly under arable farming; higher
ground to the north is under woodland and that to the south is heathland and woodland. The main
arable crops in the valley are winter cereals, potatoes, maize, turnips for winter grazing, oil seed
rape, and small areas of vegetable and salad crops. The number of dairy herds and therefore
grassland has declined in recent years. Sears (1996) notes the steady increase in cropped land as
against pasture in the Rother valley from the 1970s to the 1990s.
Figure 2 hereabouts
Lower Greensand rocks in southern England were laid down in a variety of shallow-water, near-
shore marine environments and, while dominated by sands, their textures may vary (Gallois and
Edmunds, 1965). Soil associations on these rocks have been mapped on the National Soil Map
(1:250,000) (Mackney et al., 1983) and described in more detail, including their constituent soil
series, in Jarvis et al. (1984).
Soils on the Lower Greensand outcrop in the study area are grouped as Frilford, Fyfield 1, Fyfield
2, Shirrell Heath 1 and Shirrell Heath 2 associations. The Frilford association is comprised of well-
drained sandy soils, while both Fyfield associations contain well-drained coarse to fine sandy loams
(Jarvis et al., 1984). Fyfield association soils are dominant in the four sample sites; Shirrell Heath
are mainly under wood or heathland. Within these soils there is considerable textural variability and
a lack of local data, however, Guerra (1991) sampled 81 soils from Fyfield 1 sites around Rogate
and obtained a mean value of 62% sand, 25% silt and 13% clay. Soils on eroded sites in the same
study had organic matter contents of 2.1-3.8% and bulk densities of 1.3-1.9 g cm-3. Evans (1990)
5
classifies the Fyfield 1 and 2 and the Frilford associations as high erosion risk. This assessment
takes into account current dominant land uses, landform and soil characteristics.
The propensity of sandy and loamy soils with silt and clay to crust is well documented and studies
of erosion on Lower Greensand soils have often mentioned this as a contributor to erosion (e.g.
Boardman, 1983b; Farres and Wood, 1990; Blamire, 2006). In addition, both surface and sub-
surface compaction due to arable farming also increases the risk of erosion (Jarvis et al., 1984;
Blamire, 2006). Widespread crusting was observed on bare and poorly vegetated soils in late 2006.
In the last two decades, field boundary removal has declined sharply in southern England. In the
four Sample Sites (Figure 2), a total of 21 field boundaries have been removed in comparison with
those shown on the 1997 map (Explorer 133, 1:25 000 Ordnance Survey).
The annual average rainfall at Petworth Park (1881-2007) is 857mm with highest mean monthly
amounts in December (101mm), November (99mm), October (95mm) and January (90mm). The
location of rain gauges is shown on Figure 2.
3. Erosion in the Rother valley
The Rother valley is a well known locality for soil erosion. Every winter rills and gullies can be
seen on fields adjacent to the A272 between Petersfield and Petworth with spreads of reddish sandy
soil on the road (Figure 2). Brown sandy soils around Rogate were noted as, ‘often susceptible to
quite considerable amounts of erosion…although no quantitative evidence is available’ (Nortcliff,
1978). In the early 1980s the village of Easebourne, adjacent to Midhurst, was affected by muddy
floods from arable fields. Guerra’s (1991) qualitative survey of eroding fields in the Rogate area
covered an eighteen month relatively dry period with many fields affected by wash and rilling
predominantly in the autumn and winter months. In the winter of 1994/5, Robinson (1999) carried
6
out questionnaire surveys of farmers in two erosion-prone areas of southeast England, one of them
being on the Lower Greensand between Rogate and Midhurst in the Rother valley. The erosion
hazard had little impact on land-use decisions but farmers claimed to be using low-cost anti-erosion
measures.
Sediment entering the River Lod, a tributary of the Rother, originated on post-harvest potato and
leek fields but reached the river via wheelings, a gate, a road and a drainage ditch as a the result of a
rainfall event on 1 February 1995 (6.5 mm at Petworth Park). The presence of straw bales and
sediment traps suggested that this was not the first occurrence of its kind (Theurer et al., 1998).
Sears (1996) points out the dramatic increase in high to moderate risk erosive land within 2 km of
the Rother in the period 1971-1991 and that although buffers of grass or woodland often exist along
the river, there is field evidence that these are by-passed by water and sediment using field drainage
systems, farm tracks, roads and ditches. Excessive fine sediment in the river channel is seen as a
threat to fish breeding in sand and gravel bedded rivers. A significant increase in the sediment load
entering the Rother due to the change from grazing (mainly dairy farming) to arable land use is
identified as ‘the foremost issue that currently threatens the viability and sustainability of fisheries
within the Western Rother’ (EA, 2002). The recent update of the Fisheries Action Plan for the
Rother speaks of the ‘serious threat posed to the catchment by the input of silt…impacting upon the
ecology and fish stocks of the river’ (EA, undated). Pollution from pesticides and phosphates is
also associated with siltation.
In 2002 as a response to concerns about water quality in the river, The Rother Valley Landcare
Project, a partnership between the Sussex Downs Conservation Board, English Nature and the
Environment Agency (EA), was set up to work with farmers to reduce the impact of sediment and
other pollutants on the river. The project ceased in 2005 due to loss of funding from the EA.
7
In late 2002 serious sediment pollution took place as a result of the development of a large gully
which discharged into Costers Brook, a tributary of the Rother.
In the winter of 2000-01, there was extensive erosion in the Rother valley. Air photographs taken in
March 2001 were analysed with respect to erosion features (rills, gullies and fans) formed on arable
fields. Four Sample Sites were selected for detailed analysis covering an area of 16.2 km2
(Shepheard, 2003)(Figure 2). The analysis was carried out in 2003 and therefore field checking was
not possible. The high quality coloured photographs show clearly rill networks. Erosion features on
a small number of arable fields had been ploughed-out by the time the photographs were taken. It
seems likely that all arable fields with incomplete crop cover suffered erosion during that winter
(John Blamire, personal communication). The numbers of eroded fields as seen on the air
photographs will therefore slightly underestimate actual numbers (Table 1).
In the autumn of 2006 serious erosion began in September and the four previously identified
Sample Sites were visited throughout the following year: there were 26 day-visits from 22 October
2006 to 25 October 2007. Field mapping of erosion features was carried out in these areas. Rills,
gullies and fans were recorded on base maps of about 1:15 000 scale and notes of land use, crop
cover and farming practices were made. Maps and notes were updated at every visit. Work was
concentrated on Sample Site 2 where the most serious erosion occurred. Mapping was not confined
to the boundaries of the site because fields outside the area were contributing runoff and sediment to
those within the area (Figure 3). On selected fields, measurements were made of length, cross-
sectional area and the number of rills and gullies in order to estimate erosion rates (Table 2).
Figure 3 hereabouts
8
Comparison between erosion in the two years 2000 and 2006 cannot be exact because of the
different methodologies used. However, it seems extremely likely that in both years all arable fields
with growing crops in the winter suffered erosion. The exception, noted below, is that eight fields of
ungrazed turnips in 2006 did not erode or suffered small losses (Table 2). Erosion in 2000 was not
quantified in terms of amounts. Rainfall was considerably greater in the critical months of early
winter 2000 and land use was different so that erosion was not present on all of the same fields in
the two years. Erosion was not on the same fields in the two years because the rotation of wheat,
fodder turnips, potatoes and oil seed rape is typical in the area. However, of 37 ephemeral gullies in
the four Sample Sites in 2000, 15 occurred in the same dry valleys or depressions in 2006. The
importance of ‘landscape memory’ is also emphasised by Griffin (2007).
The impact of runoff and eroded soils from fields in all Sample Sites was clearly seen with spreads
of sand on roads regularly cleared by local councils in the winter of 2006-07. The minor road
network plays an important role as a routeway for muddy runoff from fields particularly in Sample
Site 2, where many roads are eroded several metres below the level of surrounding fields (‘sunken
lanes’: Barton, 1987), with north to south gradients toward the River Rother (Figure 3). In one case,
runoff travelled over 1500m along roads to reach the Rother at Stedham Bridge. This flow, as well
as causing severe sedimentation in the river, also resulted in the flooding of a row of cottages at the
bridge. Potatoes and wheat seed from upslope fields was transported to gardens of the cottages
(BBC TV, 2007). Observations during rainfall and runoff events identified a number of points at
which muddy runoff enters the river (Figure 3). Previous studies also noted by-passing of grassed
flood plains by sediment and various points of sediment entry into the river (Sears, 1996;
Shepheard, 2003).
9
4. Assessment of risk
4.1. The Defra (2005a) scheme: context and approach
Assessment of risk of erosion has traditionally been carried out by application of one of many
available models. The assessment may refer to actual or hypothetical conditions e.g. with climate or
land-use change as a variable. The scale may be a single slope, regional, country or global and the
approach can be used to predict the impact of long-term average conditions (e.g. Universal Soil
Loss Equation) or a single event (e.g. EUROSEM). However, the trend to use complex, computer-
based models has led to issues of data availability, expert inclusion in running the models, and
validation of model results (Favis-Mortlock et al., 2001; Jetten and Favis-Mortlock, 2006;
Boardman, 2006).
The targeting of scarce public monies on sites that can be identified as at-risk of erosion has long
been advocated (Boardman, 1988a), but financial support for production on a tonnage or headage
basis has, in the European Union, been preferred. Recent introduction of agri-environmental
measures, and the principle of cross compliance, has returned the debate to how best to support
production, protect the environment, and ensure a measure of value-for-money in terms of
agricultural subsidies.
In this context, new agri-environmental measures in England and Wales propose a system of
identification of high-risk erosion sites, leading to the selection of measures to prevent erosion on
those sites, in return for financial support. Each field is assessed, by the farmer, on the basis of soil
texture and slope and a map of ranking from very high to low risk is produced for the farm (Defra,
2005a). Because farmers are asked to carry out the assessment a simple, easy-to-use system was
introduced (Table 3).
10
Erosion risk is also associated with crop cover and Defra (2005a) provides a ranking of susceptible
land uses which includes potatoes, sugar beet, maize and grazed fodder crops in the highest risk
category. A note is added that on high and very high risk sites (Table 3), the highly susceptible land
uses should be avoided unless measures are taken to control erosion. Thus, there is a clear warning
that certain combinations of soil texture, slope and crop type will lead to high risks of erosion.
Farmers are required to identify high risk sites and either put in place anti-erosion measures or
avoid high risk crops on these sites in order to meet Cross-Compliance rules and so be eligible to
receive their Single Farm Payment.
In order to test the efficacy of the Defra (2005a) system of risk assessment all fields in Sample Site
2 were assessed according to soil texture and slope (Figure 3). Texture was assessed with reference
to Table 3 and slope was estimated using the contour interval on OS 1: 25,500 scale maps. Land use
in the autumn and winter of 2006-07 was mapped (Figure 4). Fields where land use that was highly
susceptible to erosion coincided with sites classified as at high or very high risk on the basis of soil
texture and slope were recorded as non-compliant with Defra (2005a) advice (Figure 5). On these
figures sites of erosion are also represented. In the year 2006-07 measures to reduce the risk of
erosion appeared to be absent.
Figures 4 and 5 hereabouts
4.2. How successful is the Defra risk-assessment scheme?
In Sample Site 2, in the autumn of 2006, there were 25 fields with crops which are classified as
highly susceptible to erosion by Defra (2005a): winter cereals, post-harvest potatoes and grazed
fodder crops (turnips); that is all of the arable fields in the site – as listed in Table 2 (Figure 4). Of
these, 20 were on sites that are classified as at high or very high risk of erosion (Figure 3) and thus
are non-compliant with Defra advice (Figure 5). Of the 20, field mapping recorded erosion on 18.
11
Two without erosion were under turnips for grazing by sheep. This crop protected the fields from
erosion in the wetter early winter period although grazed turnips are known in this area as a high
risk crop. The ranking of fields in Table 2 shows that the most serious erosion (the top seven cases)
is on high or very high risk sites. The Defra scheme was therefore 90% successful (18/20 cases) at
identifying the high-risk sites given the land use at the time. This analysis highlights the level of
non-compliance with advice to control erosion. It is noteworthy that all cases (5) of small amounts
of erosion on sites classified at moderate risk were caused, at least in part, by runoff from upslope
fields. The Defra scheme identifies six areas of high or very high risk of erosion on steep slopes
which are kept under woodland.
4.3. Can the Defra risk-assessment scheme be improved?
In view of concern about off-site impacts of muddy runoff, including reference to the problem in
Defra (2005a), it is surprising that the assessment scheme merely considers risk on a field-by-field
basis and not the cumulative impact of runoff from a series of connected fields. In this area it is
clear that movement of runoff and soil from field to field and from field to river is a major problem
(Figure 3). Many field maps of erosion show the importance of the movement of water from field to
field, especially in situations where large blocks of land are under the same crop (Boardman, 2001;
Evans and Boardman, 2003). Risk assessment should therefore take into account the issue of
connectivity. This, however, is difficult because of the lack of information about field boundary
permeability. However, farmers are capable of assessing the passage of water and soils through
gates and field boundaries and onto roads and into rivers. Risk in such cases, can also be evaluated
after field evidence has been collected and therefore, after the event.
As a compromise, we suggest that the Defra scheme could be adapted simply by taking into account
if a high or very high risk field has a) a road downslope b) a river within 200m downslope (Walker,
2007). This will change the at-risk category of some fields and highlight those where crops
12
susceptible to erosion should be grown with care (Figure 6). On the risk of erosion map (Figure 3),
nine fields fall into the very high risk category but eight of these are under permanent grass or
woodland. On the risk of damage to road and river map (Figure 6), 19 fields are now in the very
high risk category, 12 of which are currently under permanent grassland or woodland and therefore
unlikely to generate runoff.
Figure 6 hereabouts
4.4. What is the role of rainfall?
A previous Defra scheme for erosion-risk assessment divided the country into areas above and
below a mean annual rainfall threshold of 800mm, presumably on the grounds of simplicity (Defra,
1999). This ignored the problem of rainfall variability (both spatially and temporally) and failed to
consider 'at-risk' periods where high daily / monthly totals, for example, may have coincided with
the most vulnerable periods for erosion (sowing and post-harvesting): areas that normally fall below
800mm could be at serious risk in wetter years. This is illustrated in Figure 7a.
Figure 7 hereabouts
However, a more general question remains: is there a trend towards more rain, or more intense
rainfall events that could explain increasing amounts of erosion and which should be taken into
account in any risk assessment scheme? With this in mind we have analysed the nearest long-term
rainfall record to Midhurst, that of Petworth Park, using monthly data from 1881 to 2007 and daily
data from 1907 to 2007 (Figures 7a). Four short gaps were filled with data from Midhurst which, of
local gauges, has the highest correlation with that of Petworth. We analysed changes in spring
(March - June) and autumn (September – December) rainfall because of the risk of erosion on
13
spring-sown crops (cereals, maize and potatoes) and autumn-sown crops (cereals, grazed turnips
and post-harvest fields of maize and potatoes).
The number of raindays with >10mm, >20mm, >30mm and >50mm of rain was also analysed in
view of the importance of rainfall intensity in generating runoff and erosion, especially on highly
erodible soils.
There are few statistically significant trends in the data. There is a slight decrease in the number of
>10mm raindays in spring, post-1960, and for the same period there is an increase in the number of
>20mm raindays both annually and in the autumn (Figure 7b). There is a concentration of extreme
events post-1960, with three years recording more than two >50mm daily rain falls. Two years,
2000 and 2006, were especially wet with annual totals of 1429mm (the wettest on record) and
947mm respectively. Serious erosion occurred in both years in the autumn-winter period with high
monthly totals.
There is considerable local variability in rainfall amounts even though the rain gauges in the Rother
valley are all at altitudes of between ca. 20 and 55m above sea level. This is illustrated by an
analysis of the magnitude of the rainfall event that initiated erosion in the autumn of 2006, on 28
September. For the two days, 28 and 29 September local gauges record: 26mm (Midhurst); 33mm
(Cowdray Park); 52mm (Iping Mill) and 25mm (Didling); these are all within 5km of Sample Site
2, whereas at Petworth Park, 11km distant, the total was only 15mm.
5. Control of runoff and erosion
Erosion is extensive in the Midhurst area on arable land in years such as 2000 and 2006 when
rainfall is above average. It could be argued that all land should be taken out of arable production
but this is unnecessary and unrealistic. In the four Sample Sites many fields were affected by what
14
could be defined as ‘minor erosion’. Of 54 eroding fields in 2006, only 17 contained ephemeral
gullies and rill systems. The most serious erosion was on fields in Sample Site 2 which are on high-
risk sites under crops susceptible to erosion (Table 2). Thus a limited number of fields which are
not compliant with Defra advice need be taken out of arable use or measures introduced to control
erosion. Future movement of erosion-susceptible crops onto high risk fields would have to be
avoided. Large-scale reversion to grass in the area is unrealistic because farmers are growing high-
value crops and payments under the Environmental Stewardship scheme would not compensate
them for this change. However, there is no certainty that the menus of erosion control measures
listed in Defra (2005a and 2006a) will be effective on high-risk sites, with erosion-susceptible
crops, in wetter years. To eliminate risk on such sites reversion to grass is necessary.
Measures to control muddy runoff which affects other fields, roads, watercourses and residential
properties are relevant in this area. One approach consists of detaining water within fields – the
engineering solution. Small dams or retention ponds have been used widely for this purpose e.g.
Boardman et al., 2003b; Boardman et al., 2006. In 2002, the National Trust, owners of Woolbeding
Farm, at the centre of Sample Site 2, commissioned a consultant to design a series of retention
ponds that would contain muddy runoff. This approach was adopted because the then tenant would
not consider any other measures and wished to continue farming potatoes, maize and grazed turnips
at sites that regularly eroded (Dr N Haycock, personal communication). The proposed dam-site map
shows 25 retention ponds. Mapping in the autumn of 2006 found two dams and three low,
ineffective banks and no signs of the others. However, there is no certainty that retention ponds
alone would be effective. There is considerable evidence that they can be breached, bypassed, filled
and overflow, and often are not maintained (Boardman et al., 2003b). Successful soil conservation
and flood prevention schemes require a multi-method approach combining land-use change, anti-
erosion management measures and detention structures (e.g. Stammers and Boardman, 1984; Evans
15
and Boardman 2003; Evrard et al. 2007a, 2007b). These measures are needed in a context of risk-
assessment for specific sites.
6. Discussion
The rates of erosion in Table 2 are exceptionally high in comparison with published British records
(e.g. Evans and Nortcliff, 1978; Boardman, 1988b; Boardman et al., 1996). However there is
considerable anecdotal evidence that gullying is frequent in this area, particularly where potatoes,
maize and cereals are grown and therefore the rates estimated for 2006 are not unusual. Shepheard’s
(2003) mapping of the 2000-01 events bears this out. The current paper shows clearly that the
coincidence of soil texture, slope and crop type explains the location of eroding fields. Within the
fields erosion is often localised along depressions and wheelings caused by farm vehicles (Griffin,
2007). Transfer of runoff from field-to-field is facilitated by gaps in field boundaries, loss of field
boundaries, the positioning of gates and the fact that in autumn many adjacent fields are lacking
sufficient crop cover to inhibit runoff and erosion.
The few significant upward trends in the long-term rainfall data suggest that erosion is unlikely to
be increasing as a result of changes in the pattern of rainfall, with the caveat that wet autumns such
as 2000 and 2006, undoubtedly have a significant effect. However, climate scenarios for the period
to 2080 suggest winter precipitation increases of up to 30% in southern England (Hulme et al.,
2002) and increases in the probability of very wet winters by a factor of 5 over the next 50-100
years (Palmer and Raisanen, 2002). Therefore, in the long term, risk-assessment procedures will
have to take into account these predicted changes.
It has already been noted that transfer of eroded soil from fields to rivers is facilitated by the historic
sunken lane network. Runoff from fields into sunken lanes frequently results in the cutting of bank
gullies – a process recognised in the landscape model of Farres et al. (1993) which is applicable to
16
this area (Figure 8). Runoff from fields and onto roads is frequent especially after establishment of
rills and gullies on the fields. It occurs during low magnitude events e.g. 11 December 2006 after
about 26mm on the previous two days; 21 February 2007 after about 19mm on the previous three
days.
Figure 8 hereabouts
Defra’s (2005a) risk assessment procedure has to be seen in context. It is a field-scale approach
wherein a simple question is being asked: what is the risk of erosion on a particular field? What is
not being asked, is how much erosion will occur, when it will occur and in what parts of the field.
These latter questions would require application of erosion models. Expert evaluation would be
another possible approach. Evans’ (1990) classification of 296 soil associations in England and
Wales, can be regarded as a coarse-scale scoping exercise which identifies parts of the landscape at
risk of erosion. It is supported by considerable empirical evidence of erosion rates e.g. Evans
(2005). The soil associations in this area are clearly identified as at-risk of erosion.
A further question not posed by Defra (2005a) is, what is the risk of an eroding field impacting on
roads and rivers? We suggest a simple extension of the system to take this into account.
The Defra scheme also ignores rainfall variability. Erosion on bare fields varies from year to year
depending on rainfall amount, intensity and timing. The years 2000-01 and 2006-07 were clearly
exceptional and serious erosion was widespread. However, we have already noted several other
years when erosion had a noticeable impact on roads and rivers. The increase in the arable area in
the last 30 years is likely to have increased erosion amounts and frequency.
17
The approach taken in this paper illustrates the use of simple field mapping in order to test the
validity of a risk-assessment scheme. It also highlights the various routes by which muddy runoff
from agricultural fields reaches the river. Points of entry into the river are identified (Figure 3). In
this area, the provenance of polluted runoff is easily established by mapping of rills, gullies and
runoff during and after rainfall events. However, water quality is also affected by the movement of
fine sediment from fields to rivers by routes which, in the absence of rills and gullies, are more
difficult to detect. These high-frequency, low-magnitude flows may help explain both pollution and
rising sediment yields in southern English rivers (e.g. Harrod, 1994; Foster et al., 2003; Evans,
2006).
The threat to the ecology, and in particular to the fisheries, of the Rother is clearly
acknowledged by all parties (EA, 2002, undated). The European Water Framework
Directive places responsibility, and vests powers with the Environment Agency to
return the river to ‘good ecological status’. It is not clear that the EA’s plans for
monitoring the condition of the river will be adequate nor that actions will be taken
against land managers for allowing sediment into the river.
Modifications to the Defra risk assessment scheme could be made. For example (1)
more accurate assessments of slope (2) acknowledgement of areas of steeper slopes within fields
that increase the average risk of erosion (3) inclusion of the risk of impact of runoff on roads and
rivers (4) more sophisticated assessment of runoff origin by use of a GIS flow accumulation
approach (5) some consideration of the risk of rainfall. However, all of these modifications would
mean a loss of ‘user friendliness’ and would therefore need careful consideration.
18
7. Conclusion
The success of the Defra (2005) risk-assessment scheme will depend on several factors. Firstly, that
the scheme is recognised as an efficient means of identification of at-risk sites; secondly, that it
leads to land-use change on vulnerable sites or the introduction of effective anti-erosion measures;
thirdly, that farmer compliance with recommendations is regularly checked in return for agri-
environmental payments; fourthly, that the scheme is not rendered obsolete by the recent rise in
crop prices particularly wheat; and fifthly, that it be extended to cover off-site impacts since this
would aid in the achievement of ‘good ecological status’ for rivers.
Acknowledgements
Valuable inputs into this project were made by the National Trust (Catherine Hearn), by the
Environment Agency (Michael Turner), and by consultants Dr Nick Haycock, John Blamire and
Simon Griffin. We thank the Environment Agency for access to rainfall data.
References
Barton, M.E., 1987. The sunken lanes of Southern England: Engineering geological considerations
Geological Society, London, Engineering Geology Special Publications 4, 411-417.
BBC TV, 2007. Inside Out, 3 October 2007.
Blamire, J., 2006. Soil risk assessment for Manor Farm, Woolbeding, Midhurst, West Sussex.
Unpublished report 22pp.
Boardman, J., 1983a. Soil erosion on the Lower Greensand near Hascombe, Surrey, 1982-3. Journal
Farnham Geological Society 1(3), 2-8.
Boardman, J., 1983b. Soil erosion at Albourne, West Sussex, England. Applied Geography 3, 317-
329.
Boardman, J., 1988a. Public policy and soil erosion in Britain. In: Hooke, J. M. (Ed.),
Geomorphology in Environmental Planning, John Wiley & Sons, Chichester, pp. 33-50.
19
Boardman, J., 1988b. Severe erosion on agricultural land in East Sussex, UK, October 1987. Soil
Technology 1, 333-348.
Boardman, J., 1994. Property damage by run-off from agricultural land. Town and Country
Planning 63(9), 249-251.
Boardman, J., 2001. Storms, floods and soil erosion on the South Downs, East Sussex, autumn and
winter 2000-01. Geography 86(4), 346-355.
Boardman, J., 2006. Soil erosion science: reflections on the limitations of current approaches.
Catena 68, 73-86.
Boardman, J., Evans, R., Slattery, M. C., Shuttleworth, H., 1996. Soil erosion and flooding as a
result of a summer thunderstorm in Oxfordshire and Berkshire, May 1993. Applied Geography
16(1), 21-34.
Boardman, J., Evans, R., Ford, J., 2003b. Muddy floods on the South Downs, southern England:
problem and responses. Environmental Science and Policy 6, 69-83.
Chambers, B. J., Davies, D.B., Holmes, S., 1992. Monitoring of water erosion on arable farms in
England and Wales, 1989-90. Soil Use and Management 8, 163-170.
Chambers, B. J., Garwood, T. W. D., 2000. Monitoring of water erosion on arable farms in England
and Wales. 1990-94. Soil Use and Management 16, 93-99.
Defra, 1999. Controlling Soil Erosion: A manual for the Assessment and Management of
Agricultural Land at Risk of Water Erosion in Lowland England. Department for Environment,
Food & Rural Affairs, London.
Defra, 2005a. Controlling soil erosion: a manual for the assessment and management of agricultural
land at risk of water erosion in lowland England. Revised September 2005. Department for
Environment, Food and Rural Affairs, London.
Defra. 2005b. Entry level stewardship handbook: terms and conditions and how to apply, Rural
Payments Agency and Department for Environment, Food and Rural affairs, London.
20
Defra, 2006a. Single payment scheme: cross compliance guidance for soil management, 2006
edition, Rural Payments Agency and Department for Environment, Food and Rural affairs,
London.
Defra, 2006b. Single payment scheme: cross compliance handbook for England 2006 edition, Rural
Payments Agency and Department for Environment, Food and Rural affairs, London.
EA, 2002. Western Rother Fisheries Action Plan, Environment Agency.
EA, undated. Western Rother Fisheries Action Plan, Environment Agency.
Evans, R., 1971. The need for soil conservation. Area 3(1), 20-23.
Evans, R., 1990. Soils at risk of accelerated erosion in England and Wales. Soil Use and
Management 6, 125-131.
Evans, R., 1993. Extent, frequency and rates of rilling of arable land in localities in England and
Wales. In: Wicherek. S. (Ed.), Farm Land Erosion: In Temperate Plains Environment and Hills,
Elsevier, pp. 177-190.
Evans, R., 2005. Monitoring water erosion in lowland England and Wales – a personal view of its
history and outcomes. Catena 64, 142-161.
Evans, R., 2006. Land use, sediment delivery and sediment yield in England and Wales. In: Owens,
P.N., Collins, A.J. (Eds.), Soil Erosion and Sediment Redistribution in River Catchments, CAB
International, Wallingford, UK, pp. 70-84.
Evans, R., Boardman, J 2003. The curtailment of muddy floods in the Sompting catchment, South
Downs, West Sussex, southern England. Soil Use and Management 19, 223-231.
Evans, R., Nortcliff, S., 1978. Soil erosion in north Norfolk. Journal of Agricultural Science,
Cambridge 90, 185-192.
Evrard, O., Persoons, E., Vandaele, K., van Wesemael, B., 2007a. Effectiveness of erosion
mitigation measures to prevent muddy floods: a case study in the Belgian loam belt. Agriculture,
Ecosystems and Environment 118, 149-158.
21
Evrard, O., Bielders, C.L., Vandaele, K., van Wesemael, B., 2007b. Spatial and temporal variation
of muddy floods in central Belgium, off-site impacts and potential control measures. Catena 70,
443-454.
Farres, P.J., Wood, S., 1990. Some characteristic soil erosion landscapes of N.W Europe. South
Hampshire Geographer 19, 38-44.
Farres, P., Poesen, J., Wood, S., 1993. Soil erosion landscapes. Geography Review 6(3), 38-41.
Favis-Mortlock, D., Boardman, J., MacMillan, V. J., 2001. The limits of modelling: why we should
proceed with care. In: Harmon, R., Doe, W.W. (Eds.), Landscape Erosion and Evolution
Modelling, Kluwer Academic/Plenum Publishers, pp. 477-516.
Foster, I.D.L., Chapman, A.S., Hodgkinson, R.M., Jones, A.R., Lees, J.A., Turner, S.E., Scott, M.
2003. Changing suspended sediment load and particulate phosphorus loads and pathways in
underdrained agricultural catchments: Herefordshire and Worcestershire, UK. Hydrobiologia
494, 119-126.
Gallois, R. W., Edmunds, F. H., 1965. British Regional Geology: The Wealden District (Fourth
Edition), HMSO, London.
Griffin, S., 2007. A spatial and temporal study of soil erosion in West Sussex from 2001-2006.
Unpublished paper, South East England Soils Discussion Group,
Guerra, A. J. T., 1991. Soil characteristics and erosion, with particular reference to organic matter.
Unpublished PhD thesis, King’s College, University of London.
Holman, I.P., Hollis, J.M., Bramley, M.E., Thompson, T.R.E. 2003. The contribution of soil
structural degradation to catchment flooding: a preliminary investigation of the 2000 floods in
England and Wales. Hydrology and Earth System Science 7(5), 754-765.
Hulme, M., Jenkins, G.J., Lu, X., Turnpenny, J., Mitchell, T. D., Jones, R. G., Lowe, J., Murphy,
J. M., Hassell, D., Boorman, P., McDonald, R, Hill, S. 2002. Climate change scenarios for the
United Kingdom: the UKCIP02 scientific report, Tyndall Centre for Climate Change Research,
School of Environmental Sciences, University of East Anglia, Norwich.
22
Jackson, S. J., 1986. Soil erosion survey and soil erosion risk in mid-Bedfordshire. SEESOIL 3, 95-
104.
Jarvis, M. G., Allen, R.H., Fordham, S.J., Hazelden, J., Moffat, A.J., Sturdy, R.G., 1984. Soils and
their use in South East England. Soil Survey of England and Wales, Harpenden.
Jetten, V., Favis-Mortlock, D. 2006.Modelling soil erosion in Europe. In: Boardman, J., Poesen, J.
(Eds.), Soil Erosion in Europe, John Wiley & Sons, Chichester, pp. 696-716.
Mackney , D., Hodgson, J.M., Hollis., J.M., Staines, S.J., 1983. Legend for the 1:250,000 Soil
Map of England and Wales. Soil Survey of England and Wales, Harpenden.
Morgan, R. P. C., 1977. Soil erosion in the United Kingdom: field studies in the Silsoe area, 1973-
75. Occasional Paper No. 4, National College of Agricultural Engineering, Cranfield Institute of
Technology.
Morgan, R. P. C., Martin, L., Noble, C. A. 1987. Soil erosion in the United Kingdom: a case study
from mid-Bedfordshire. Occasional Paper No. 14, Silsoe College, Cranfield Institute of
Technology, Silsoe, Bedfordshire.
Morgan, R. P. C., Quinton, J. N., Smith, R. E., Govers, G., Poesen, J. W. A., Auerswald, K.,
Chischi, G., Torri, D., Styczen, M.E., 1998. The European Soil Erosion Model (EUROSEM): a
dynamic approach for predicting sediment transport from fields and small catchments. Earth
Surface Processes and Landforms 23, 527-544.
Nortcliff, S., 1978. Soils of the Rogate area. Rogate Papers 2, King’s College London, Rogate Field
Centre, 29pp.
Palmer, T. N. Raisanen, J. 2002 .Quantifying the risk of extreme seasonal precipitation events in a
changing climate. Nature 415, 512–14.
Quinton, J. N., Catt, J.A., 2004. The effects of minimal tillage and contour cultivation on surface
runoff, soil loss and crop yield in the long-term Woburn Erosion Reference Experiment on sandy
soils at Woburn, England. Soil Use and Management 20, 343-349.
23
Robinson, D.A., 1999. Agricultural practice, climate change and the soil erosion hazard in parts of
southeast England. Applied Geography 19(1), 13-27.
Scaife, R.G., Macphail, R.I., 1983. The post-Devensian development of heathland soils and
vegetation. SEESOIL 1, 70-99.
Sears, D. A., 1996. Fine sediment accumulations in the River Rother, West Sussex. Report to
Southern Region NRA.
Shepheard, M.L. 2003. Soil erosion and off-site impacts from the Lower Greensand soils and arable
lands of the Rother river valley. Unpublished MSc dissertation, Environmental Change Institute,
University of Oxford.
Stammers, R., Boardman, J. 1984. Soil erosion and flooding on downland areas. The Surveyor 164,
8-11.
Theurer, F.D., Harrod, T.R., Theurer, M., 1998. Sedimentation and salmonids in England and
Wales, R&D Technical Report P194, Environment Agency.
Turnbull, A., 1985. An investigation into soil erosion by water in the Arreton valley – Apse Heath –
Princelett area of the Isle of Wight. Unpublished BSc dissertation, Faculty of Natural and Life
Sciences, Brighton Polytechnic.
Walker, E., 2007. Soil erosion and Defra’s Catchment Sensitive Farming Programme – a
Herefordshire case study. Unpublished MSc dissertation, Environmental Change Institute,
University of Oxford.
Table 1: Erosion in the Sample Sites
Sample Site Area (km2 ) Number eroding fields
2000
Number eroding fields
2006 1 3.95 17 13 2 4.90 19 21 3 2.58 7 3 4 4.77 12 17
Total 16.20 55 54
24
Table 2: Soil loss and erosion risk (Defra) on arable fields in Sample Site 2, winter 2006-07. Fields are ranked semi-quantitatively based on soil loss from top (most serious) to bottom of the Table. (Field numbering as on Figure 4). Field no. and
area (ha) Land use Erosion forms Soil loss
(m3 ha-1) Defra
erosion risk 10 (7.2) post-harvest
potatoes rills, gullies, fans
180* high
21 (7.5) winter cereal rills, fans 91* (383**) high (very high)
16 (4.1) winter cereal rills, gullies, fans
56* (231**) high (very high)
26 (9.4) winter cereal rills, gullies, fans
50* high
8 (9.9) winter cereal rills, gullies, fans
49* high
11 (4.3)
winter cereal rills, gullies, fans
31* very high
9 (8.9) post-harvest potatoes
rills, gullies, fans
24* high
24 (9.2) post-harvest potatoes
small gully,+ few rills
nm moderate
22 (13.8) winter cereal rills, fans a small gully+
nm high
39 (10.8) winter cereal rills & small gully
nm high
4 (12.2) winter wheat rills and small gully
nm high
59 & 36 (17.6) winter cereal small gullies and rills
nm high
20 (4.1) winter cereal rills and fans+ nm high 33 (6.3) winter cereal small gullies nm high 38 (5.4) winter cereal gully and fan nm high 32 (7.4) forage turnips small gully+ nm moderate 12 (6.0) forage turnips small gully+ nm moderate 13 (6.4) forage turnips small gully+ nm moderate 7 (11.8) forage turnips small gully+ nm moderate 27 (8.9) forage turnips small gully nm high 2 (8.3) forage turnips small gully nm high 34 (4.6) winter cereal small gully and
fan nm high
31 (15.9) forage turnips no erosion / high 1 (2.8) forage turnips no erosion / high
*average rate for whole field; ** average for worst affected part of field; nm not measured; +mainly due to runoff from higher fields.
25
Table 3: Water erosion risk assessment
Soils Steep slopes > 70
Moderate slopes 30 - 70
Gentle slopes 20 - 30
Level ground < 20
Sandy and light silty soils
Very high High Moderate Lower
Medium and calcareous soils
High Moderate Lower Lower
Heavy soils
Lower Lower Lower Lower
From Defra (2005a): Controlling Soil Erosion
Figure 1: The Lower Greensand outcrop in southern England
26
Figure 2: Location of Sample Sites in the Rother Valley Rain gauges: a. Iping Mill; b. Didling; c. Cowdray Park; d. Midhurst; e. Petworth Park
Figure 3: Risk of erosion in Sample Site 2 based on soil texture and slope
27
Figure 4: Land use in Sample Site 2 autumn and winter 2006-07
Figure 5: Compliance with Defra (2005a) advice based on soil texture, slope and land use, autumn and winter 2006-07 in Sample Site 2
28
Figure 6: Risk of damage to road or river downslope of field in Sample Site 2
Figure 7a. Annual rainfall Petworth Park 1881-2007
0
200
400
600
800
1000
1200
1400
1600
1880 1890 1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000 2010
Rain
fall
(mm
)
AnnAnnMovAv
29
Figure 7b. The number of >20mm raindays in autumn-winter at Petworth Park post-1960
0.010.020.030.040.050.060.070.080.090.0
100.0
1950 1960 1970 1980 1990 2000 2010
Figure 8: Erosion on agricultural landscapes in northwest Europe (Farres et al., 1993)
top related